US 20140054231A1
(19) United States
SPENCE et al. (43) Pub. Date: Feb. 27, 2014
JIM LORENTZ, Fort McMurray (CA);
Publication Classi?cation
OIL SANDS TAILINGS
(75) Inventors: JONATHAN SPENCE, Edmonton
(51) Int. Cl.
B01D 21/01
(CA); BARRY BARA, Edmonton (CA); (52) US, Cl,
(2006.01)
USPC ......... .. 210/710; 210/702; 210/726; 210/738;
210/729; 210/734
JIWON LEE, Fort McMurray (CA);
RICHARD DANIEL LAHAIE, (57) ABSTRACT
A
. . .. . . process for deWatermg o1l sands ta1l1ngs 1s provlded, com prising providing a tailings feed having a solids content in the for the owners of syncrude Proj ect’
Fort MCMurray (CA)
(21) App1_ NO_; 13/594,402 to the tailings feed and mixing the tailings feed and ?occulant to form ?ocs; and centrifuging the ?occulated tailings feed to produce a centrate having a solids content of less than about
(22) Filed: Aug. 24, 2012 Wt %.
Bypass
6%
Optional Comma Qilu?orURecwe
.. .. . s age v r
‘ W
'
{?ring xx: “W99
Tanks giixets
{if lfliii
3-4 40 5g 44 >
/ . t i j; i i 5 E2; r 5 . >
4:
: -
. i I Rgii?ie? Fees‘ mews i ;
3; ;
I 42
E
46 ~
D C
£232
1" T "f Y
5
. "
“"1”;
Pigment
Mewng
AB
5
5 t
5‘
' l _
" Cemrate
W?iiéf Tank
Effluent in Pond

Patent Application Publication Feb. 27, 2014 Sheet 1 0f 11 US 2014/0054231 A1

Patent Application Publication Feb. 27, 2014 Sheet 2 0f 11 US 2014/0054231 A1
‘NC?
T:
“5 3!} w
U
ZGJWi-I I
| |
10-Aug-11 25-A11g-1 I
|
U9-Sap—'1 1
|
Z‘lé-fieg- I 1 DWQCYJ 1
gm
5U
% 5O g
Q in}
331 i" 30
‘
i
_)
A v v g
""2 f g‘ I
Y L g t f ",1
$i2e in lviim'mns
Monc?m-imge
& MLEB All S
4
A MLSB EFT
.,
FIG. 3

Patent Application Publication Feb. 27, 2014 Sheet 4 0f 11 US 2014/0054231 Al l l l I l l l l l
FIG. 6
8
4
U
0
I
0/
0,05 0.1 0,15 02 ‘0.25
FIG. 7
0.3 0.35

Patent Application Publication Feb. 27, 2014 Sheet 5 0f 11 US 2014/0054231 A1
7;“: 45445.‘;2x - 13.122
R. = 0.9219
L’T (degrees K}
FIG. 8
O
{3.12 (L14 0.16 {3.1.3 (3&2 0,12 0,14 0J6 0,2‘ 0.3 MUN:
Corrected polymm‘ mncmtraticm
FIG. 9

Patent Application Publication Feb. 27, 2014 Sheet 6 0f 11 US 2014/0054231 A1
120
40
0212 {3.13 0.14 0.15 0.16 0.17 0.18 0.19 0.2 0.21 0.22 {1.23 0.2.4 M0112
(‘oriental polymer cmlcea'nmtion
FIG. 10
O
H)
(2.2 0.22 0.24 0.26 0.28 0.3 0.32 0.34 0.36 (L33 0.4 0.42 0.44 0.46 0.48 (3.5
CDYTECEEii pnlymer cmimcentn'atiou
FIG. 11

Patent Application Publication Feb. 27, 2014 Sheet 7 0f 11
0/ Fims Ciaptua'e
{ii-nary Cnpmre
Salida Capturc
FIG. 12
‘b Feed
Cimrmte
, 1 (Cake
Particle Size in Microns

Patent Application Publication Feb. 27, 2014 Sheet 8 0f 11 US 2014/0054231 A1
MC} g if;
%
W a i
M1 a»
4!":51 i‘ w ' 5g ,3‘
91 “1'- i 2011 L5-T1X1000
|
l
,
0
I
0
*
I.
‘ ' l
‘if n i
|
‘
‘k
i z "
FIG. 14

Patent Application Publication Feb. 27, 2014 Sheet 9 0f 11 US 2014/0054231 A1
FIG. 15

Patent Application Publication Feb. 27, 2014 Sheet 10 0f 11 US 2014/0054231 A1
1109 i 1 i l
‘ 109
- 90
1900 m “m
- 80 Q“
(‘3 w I
as
MM‘ _
800 gimme ?occulant
0 | 5
831111315: Number
6f’ if, a» 0 4 0' ‘I, Q ' 49 kf
25
FIG. 16

Patent Application Publication Feb. 27, 2014 Sheet 11 0f 11 US 2014/0054231 A1
32 v,
‘9% 30 ~
5;’
%
28 -
'34 —
22
1» 1,9 micron ~* Polymea'
* 25%
4 _ g
‘if’ an
~ 15% ,g
M 501:?
20 I I I I I I I I I
0
Mmxth and Day I
"8
$ m T;
u
Q1
0 v 55 micmn - Paiymer
25W a.)
24300 is‘
I 1500 .g
5; ' 1W0 a;
Q.‘
~- 500
25
I I | I I I I I I
U

US 2014/0054231 A1 Feb. 27, 2014
FIELD OF THE INVENTION
[0001] The present invention relates to a process for deWa tering oil sands tailings. In particular, tailings are treated With a coagulant and a ?occulant and subjected to centrifugation to desiccation.
OIL SANDS TAILINGS
[0010] (5) ultra?nes separation does not occur With ?occu lated centrifuge feed. Surprisingly, the particle siZe distribu tion did not differ among the centrifuge feed, cake and cen trate;
[0011] (6) it Was surprisingly discovered that the process
Worked at ambient temperature; and
[0012] (7) the optional addition of a coagulant may result in higher throughput and produce a signi?cantly stronger, more conveyable cake from the centrifuge.
[0013] Thus, use of the present invention enables reclama tion of tailings disposal areas and recovers Water suitable for recycling in the process.
[0014] In one aspect, a process for deWatering oil sands
[0015] providing a tailings feed having a solids content
BACKGROUND OF THE INVENTION
[0002] Oil sand generally comprises Water-Wet sand grains held together by a matrix of viscous heavy oil or bitumen.
Bitumen is a complex and viscous mixture of large or heavy hydrocarbon molecules Which contain a signi?cant amount of sulfur, nitrogen and oxygen. The extraction of bitumen from sand using hot Water processes yields large volumes of ?ne tailings composed of ?ne silts, clays, residual bitumen and
Water. Mineral fractions With a particle diameter less than 44 microns are referred to as “?nes.” These ?nes are typically clay mineral suspensions, predominantly kaolinite and illite.
[0003] The ?ne tailings suspension is typically 85% Water and 15% ?ne particles by mass. DeWatering of ?ne tailings occurs very sloWly. When ?rst discharged in ponds, the very loW density material is referred to as thin ?ne tailings. After a feW years When the ?ne tailings have reached a solids content of about 30-35%, they are referred to as ?uid ?ne tailings
Which behave as a ?uid-like colloidal material. The fact that
?uid ?ne tailings behave as a ?uid and have very sloW con solidation rates signi?cantly limits options to reclaim tailings
Water from the ?uid ?ne tailings to strengthen the deposits so that they can be reclaimed and no longer require containment.
[0004] Accordingly, there is a need for an improved method to treat ?ne tailings to reduce their Water content and reclaim the land on Which ?ne tailings are disposed.
[0016] adding a ?occulant to the tailings feed and mixing the ?occulant and tailings feed to form ?ocs; and
[0017] centrifuging the ?occulated feed to produce a and a cake having a solids content of at least about 50 Wt
%.
[0018] In one embodiment, a coagulant is added to the tailings feed prior to centrifugation. In another embodiment, a coagulant is added to the tailings feed prior to the addition of the ?occulant.
[0019] In one embodiment, the oil sands tailings is ?uid ?ne tailings, Which ?uid ?nd tailings may be optionally diluted
With Water to provide the tailings feed having a solids content embodiment, the tailings feed has a solids content in the range
[0005] The current application is directed to a process for deWatering oil sands tailings by treating the tailings With coagulant and ?occulant prior to deWatering by centrifuga tion. The present invention is particularly use?ll With, but not limited to, ?uid ?ne tailings. It Was surprisingly discovered that by conducting the process of the present invention, one or more of the folloWing bene?ts may be realiZed:
[0006] (1) providing a concentrated ?occulant solution may reduce the volume of high quality ?occulant make up
Water Which Would normally be required, and corresponds
[0007] (2) the ?occulant may be mixed With tailings having a solids content of greater than 30 Wt %, thus minimiZing the
[0008] (3) optimum mixing of the ?occulant and tailings may be achieved by injecting the ?occulant at a point directly before the centrifuge feed tube to avoid overshearing;
[0009] (4) deWatering by centrifugation may produce a cen trate having a solids content of less than about 3 Wt %, and a capturing greater than 95% of the solids Within the initial
[0020] In one embodiment, the ?occulant is a Water soluble polymer having a moderate to high molecular Weight and an intrinsic viscosity of at least 3 dl/g (measured in 1N NaCl at
250 C.).
BRIEF DESCRIPTION OF THE DRAWINGS
[0021] Referring to the draWings Wherein like reference numerals indicate similar parts throughout the several vieWs, several aspects of the present invention are illustrated by Way of example, and not by Way of limitation, in detail in the
?gures, Wherein:
[0022] FIG. 1 is a schematic of one embodiment of the present invention for treating oil sands tailings prior to deWa
[0023] FIG. 2 is a graph shoWing the consistency in the solids content (Wt %) of the ?uid ?ne tailings from the dredge.
[0024] FIG. 3 is a graph shoWing the average mineral par ticle siZe distribution of four samples of ?uid ?ne tailings.
[0025] FIG. 4 is a graph shoWing the average 44 micron fraction in the ?uid ?ne tailings.
[0026] FIG. 5 is a graph shoWing the average 5.5 micron fraction in the ?uid ?ne tailings.
[0027] FIG. 6 is a graph shoWing the average 1.9 micron fraction in the ?uid ?ne tailings.
[0028] FIG. 7 is a graph shoWing the relationship betWeen polymer viscosity and concentration for the ?occulant at 18°
C. using a simple constant rpm rheometer (Fann model at 200
[0029] FIG. 8 is a graph shoWing the Arrhenius relationship for the 0.2% polymer solution.

US 2014/0054231 A1 Feb. 27, 2014
[0030] FIG. 9 is a histogram showing a summary of all the nominally 0.2% polymer tests indicating that most of the polymer is Within 10% of the target concentration.
[0031] FIG. 10 is a histogram shoWing polymer concentra tions With intercept corrected data bringing the average to
0.2%.
[0032] FIG. 11 is a histogram shoWing a summary of the temperature corrected 0.4% polymer concentrations using the slope from the 0.2% Arrhenius data With intercept cor rected for a 0.4% average concentration.
[0033] FIG. 12 is a graph comparing the ?nes and clay capture to solids capture for all the experimental runs.
[0034] FIG. 13 is a graph shoWing the results of Coulter particle siZe analysis for centrifuge feed, centrate, and cake samples over the entire testing.
[0035] FIG. 14 is a graph shoWing the relationship betWeen centrate solids and solids capture for the three pilot tests.
[0036] FIG. 15 is a graph shoWing centrate solids content as a function of throughput during the high capacity test, With the inset shoWing the rapidly settling centrate.
[0037] FIG. 16 is a graph shoWing the solids capture (%) in a 24 hour loW polymer dosage test.
[0038] FIG. 17 illustrates tWo graphs shoWing the general trend betWeen polymer dosage and clay content for 1.9 micron clay particles (A) and 5.5 micron clay particles (B).
DETAILED DESCRIPTION OF THE PREFERRED
EMBODIMENTS
[0039] The detailed description set forth beloW in connec tion With the appended draWings is intended as a description of various embodiments of the present invention and is not intended to represent the only embodiments contemplated by the inventor. The detailed description includes speci?c details for the purpose of providing a comprehensive understanding of the present invention. HoWever, it Will be apparent to those skilled in the art that the present invention may be practiced
Without these speci?c details.
[0040] The present invention relates generally to a process for treating tailings derived from oil sands extraction opera tions and containing a ?nes fraction, and deWatering the tail ings to enable reclamation of tailings disposal areas and to recover Water for recycling. As used herein, the term “tail ings” means tailings derived from oil sands extraction opera tions and containing a ?nes fraction. The term is meant to include ?uid ?ne tailings (FFT) from tailings ponds and ?ne
thickener under?oW or froth treatment tailings) Which may bypass a tailings pond. The tailings are treated With coagulant and ?occulant prior to deWatering by centrifugation to aggre gate the solids and to recover the Water.
[0041] FIG. 1 is a ?oW diagram ofthe process ofthe present invention. In one embodiment, the tailings are primarily FFT obtained from tailings ponds. HoWever, it should be under stood that the ?ne tailings treated according the process of the present invention are not necessarily obtained from a tailings pond and may also be obtained from ongoing oil sands extrac tion operations.
[0042] The tailings stream from bitumen extraction is typi cally transferred to a tailings pond 10 Where the tailings stream separates into an upper Water layer, a middle FFT layer, and a bottom layer of settled solids. The FFT layer 12 is removed from betWeen the Water layer and solids layer via a dredge 14 or ?oating barge having a submersible pump. In one embodiment, the FFT 12 has a solids content ranging the FFT 12 has a solids content ranging from about 30 Wt % to about 45 Wt %. In one embodiment, the FFT 12 has a solids
FFT 12 is preferably undiluted. The FFT is passed through a screen 16 to remove any oversiZed materials. The screened
FFT 12 is collected in a vessel such as a tank 18. In one embodiment the FFT 12 is then pumped via a pump 20 from the tank 18 into an agitated feed tank 22 comprising a tank body and blades. In another embodiment FFT is pumped to a simple surge tank., and in yet another embodiment FFT is pumped directly to the centrifuge.
FFT prior to entering the agitated feed tank 22. In one embodiment, coagulant 24 is introduced into the in-line ?oW of FFT prior to entering the centrifuge 38. As used herein, the term “coagulant” refers to a reagent Which neutraliZes repul sive electrical charges surrounding particles to destabiliZe suspended solids and to cause the solids to agglomerate.
Suitable coagulants include, but are not limited to, gypsum, lime, alum, polyacrylamide, or any combination thereof. In one embodiment, the coagulant comprises gypsum or lime.
As used herein, the term “in-line ?oW” means a ?oW con tained Within a continuous ?uid transportation line such as a pipe or another ?uid transport structure Which preferably has an enclosed tubular construction. Su?icient coagulant 24 is added at line 26 to initiate neutralization. The dosage of the coagulant 24 is controlled by a metering pump 28. In one embodiment, the dosage of the coagulant 24 ranges from about 300 grams to about 1,500 grams per tonne of solids in the FFT.
[0044] Dilution Water 30 is required to disperse the coagu lant 24 into the forWard ?oW of the FFT 12 and to minimiZe the risk of total coagulation Which Would entrap the solids
Within the line 26. The dilution Water 30 is introduced into the in-line ?oW of the FFT at line 26 prior to entering the agitated feed tank 22. The source of Water 30 is preferably any loW solids content process affected Water. The FFT 12 and diluted coagulant 24 are blended together Within the agitated feed tank 22, or in the pipeline When no feed tank is used. Agitation is conducted for a su?icient duration in order to alloW the coagulant 24 to dissociate from the Water 30 and agglomerate the FFT 12. In one embodiment, the duration is at least about
?ve minutes.
[0045] The agitated FFT 34 is then diluted With Water 30.
The Water 30 is introduced into the in-line ?oW of the agitated
FFT 34 prior to entering a mixer 44 . As previously mentioned, the source of Water 30 is preferably any loW solids content process affected Water. Su?icient Water 30 is added to achieve a centrifuge feed 36 having a solids content preferably in the than about 30 Wt %. Dilution provides a consistent feed 36 to the centrifuge 38 to ensure stable machine operation. In one embodiment, the diluted FFT 40 is pumped via a pump 42 from the agitated feed tank 22 into the mixer 44. In another embodiment FFT is piped directly to the mixer 44.
[0046] Additional Water 30 and a ?occulant 46 are intro duced into the in-line ?oW of the diluted FFT 40 at a line 54 prior to entering the mixer 44. As used herein, the term “?oc culant” refers to a reagent Which bridges the neutraliZed or
more e?icient settling. Flocculants useful in the present invention are generally anionic, nonionic, cationic or ampho teric polymers, Which may be naturally occurring or syn

US 2014/0054231 A1 Feb. 27, 2014
the polymeric ?occulants are characterized by molecular
Weights ranging betWeen about 1,000 kD to about 50,000 kD.
Suitable natural polymeric ?occulants may be polysaccha rides such as dextran, starch or guar gum. Suitable synthetic polymeric ?occulants include, but are not limited to, charged or uncharged polyacrylamides, for example, a high molecular
[0047] Other useful polymeric ?occulants can be made by
mide and polyethylene glycol methacrylate, and one or more anionic monomer(s) such as acrylic acid, methacrylic acid,
2-acrylamido-2-methylpropane sulphonic acid (ATBS) and salts thereof, or one or more cationic monomer(s) such as
chloride (DADMAC), acrylamido propyltrimethyl ammo nium chloride (APTAC) and/ or methacrylamido propyltrim ethyl ammonium chloride (MAPTAC).
[0048] In one embodiment, the ?occulant 46 comprises an aqueous solution of an anionic polyacrylamide. The anionic polyacrylamide preferably has a relatively high molecular
Weight (about 10,000 kD or higher) and medium charge den sity (about 20-35% anionicity), for example, a high molecular
The preferred ?occulant may be selected according to the
FFT composition and process conditions.
[0049] The ?occulant 46 is supplied from a ?occulant make up system for preparing, hydrating and dosing of the ?occu lant 46. Flocculant make-up systems are Well knoWn in the art, and typically include a polymer preparation skid 48, one or more storage tanks 50, and a dosing pump 52. The dosage of ?occulant 46 is controlled by a metering pump 56. In one embodiment, the dosage of ?occulant 46 ranges from about
400 grams to about 1,500 grams per tonne of solids in the FFT.
In one embodiment, the ?occulant is in the form of a 0.4% solution.
[0050] The additional Water 30 is provided to disperse the
?occulant 46 into the forWard How of the diluted FFT 40 and to minimiZe the risk of total ?occulation Which Would entrap the solids Within the line 54. When the ?occulent 46 contacts the diluted FFT 40, it starts to react to form ?ocs formed of multiple chain structures and PET minerals. The diluted FFT
40 and diluted ?occulant 46 are further combined Within the mixer 44. Since ?occulated material is shear-sensitive, it must be mixed in a manner so as to avoid overshearing. Over shearing is a condition in Which additional energy has been input into the ?occulated FFT, resulting in release and re suspension of the ?nes Within the Water. Suitable mixers 44 include, but are not limited to, T mixers, static mixers, dynamic mixers, and continuous-?ow stirred-tank reactors.
Preferably, the mixer 44 is a T mixer positioned before the feed tube (not shoWn) of the centrifuge 38. In one embodi ment, diluted ?occulant 46 may bypass the mixer (44) and be fed directly to the feed line of the centrifuge 38 for addition to the diluted FFT 40.
[0051] Flocculation produces a suitable feed 36 Which can be deWatered in the centrifuge 38. The feed 36 is transferred to the centrifuge 38 for deWatering. In one embodiment, the centrifuge 38 is a solid boWl decanter centrifuge. Solid boWl decanter centrifuges are capable of deWatering materials
Which are too ?ne for effective deWatering by screen boWl centrifuges. Extraction of centrate 58 occurs in the cylindrical part of the boWl, While deWatering of solids by compression of the cake 60 takes place in the conical part of the boWl.
Separation of the centrate 58 and cake 60 using a solid boWl decanter centrifuge may be optimally achieved using loW
and high boWl speed rpm.
[0052] In one embodiment, the centrate 58 has a solids content of less than about 3 Wt %. The centrate 58 may be collected into a tank 62 and either discharged back to the tailings pond 10, or diverted into a line 64 for recycling for
?occulant make-up or feed dilution.
[0053] In one embodiment, the cake 60 has a solids content of at least about 50 Wt %. The cake 60 may be collected and transported via a conveyor 66, pump or transport truck to a disposal area 68. At the disposal area 68, the cake 60 is stacked to maximiZe deWatering by natural processes includ ing consolidation, desiccation and freeZe thaW via 1 to 2 m thick annual lifts to deliver a tra?icable surface that can be reclaimed. In another embodiment, cake can be placed in deep pits Where deWatering includes desiccation and freeZe thaW, but primarily consolidation. In another embodiment, cake is placed at the bottom of End Pit Lakes.
[0054] Exemplary embodiments of the present invention are described in the folloWing Example, Which is set forth to aid in the understanding of the invention, and should not be construed to limit in any Way the scope of the invention as de?ned in the claims Which folloW thereafter.
Example 1
[0055] PET was obtained from an oil sand tailings settling basin using a Royal Boskalis Westminster type IHC 1500 cutter suction dredger capable of pumping 1900 m3/hr of PET and obtaining FFT from levels as deep as 11 meters doWn in the pond. Dredged PET was pumped to the testing site, and screened through a 3A><3A inch ?xed screen prior to entering the feed tank. The FFT supply system Was run continuously. cess affected Water and environmental run-off Water from a series of ponds at the base of the dike. The chemistry of the
Water is set out in Table 1.
TABLE 1
Cation
Concentration
(ppm)
Anion
Concentration (ppm) Other
Ca Mg Na Cl S04 HCO3 CO3 pH Ion Balance
12 4 444 210 77 720 41 8.47 0.98
[0057] A ?occulant make-up skid (SNF Floerger, France)
Was used to prepare a ?occulant solution. 750 kg bags of polyacrylamide polymer (SNF Flopam 3338) Were made up to a mother liquor concentration of 1.5% by Weight and diluted to a concentration of either 0.2% or 0.4% using pro cess affected and environmental run-offWater, and stored in a
60 m3 storage tank until use. In one embodiment, the ?occu lant is an acrylamide-acrylate copolymer. In another embodi ment, the ?occulant is a high molecular Weight (e.g., 14-20
[0058] A gypsum supply system provided gypsum slurry.
Agricultural grade gypsum needs about 7 minutes to dissolve

US 2014/0054231 A1 Feb. 27, 2014 properly in an FFT slurry. At feed rates in excess of 100 m3/h, the 30 m3 FFT storage tank provided about 20 minutes of residence time for the gypsum to go into solution. The gyp sum slurry Was nominally made up to 2% solids by Weight, and added via a metering pump to the FFT line.
[0059] PET was pumped from the feed tank to individual agitated feed tanks, With each tank provided With a commer cially available centrifuge. In this example, an Alfa Laval
Lynx 1000 Was used. When used, gypsum Was added to the
FFT prior to each agitated feed tank. Flocculant solution Was added to the feed after the agitated feed tanks. Mixing of the
PET and diluted ?occulant Was tested using a simple T mixer, static mixer and a continuous-?ow stirred-tank reactor. Sat isfactory mixing Was achieved With the T mixer positioned directly before the centrifuge feed tube.
[0060] The centrifuges Were operated in parallel. The Alfa
Laval centrifuge Was provided With tWo rotating assemblies,
With rotating assembly #2 having shalloWer beach angle. The initial gear box installed on the Alfa Laval centrifuge pro vided a limited back drive capability, Which Was subsequently improved to alloW more back drive capacity.
[0061] Following centrifugation, the cake Was collected via a conveyor, and transferred to a single open ended discharge cell. The production rate of cake Was measured from each centrifuge using bins on load cells. Cake rates Were measured for key test conditions to con?rm material balances. Centrate from each centrifuge Was dropped into separate collection tanks, and the ?nal centrate Was pumped back to the Mildred
Lake Settling Basin.
[0062] Each of the key process lines Was equipped to alloW sampling. FloW and density meters Were installed for process control and mass balancing. Magnetic ?oW meters (Endress rheometer or a Fann constant RPM viscometer operated at
200 rpm. Centrifuge cake rheology Was determined using a
Haake V1scotester 550.
[0064] Oil/Water/ solids composition Was determined using
Sedigraph III 5120). For Water chemistry, the pH, bicarbonate and carbonate concentrations Were determined With a PC
Titrate Alkalinity Autotitrator (Mandel); elemental analysis using a Varian Simultaneous Vista-Pro ICP-OES; and anions using a Dionex ICS 3000.
[0065] In addition, or, in the alternative, oil/Water/ solids content Was determined With a Dean Stark soxhlet extraction technique With hot toluene. Large extractors Were used for the centrate, and small extractors Were used for the FFT, centri fuge feed, and cake. The particle siZe distributions of hydro carbon free solids Were measured With the Coulter Particle
Analysis technique, using a Coulter LS 13 320 laser diffrac tion particle analyZer. The solids Were cleaned using the Dean
& Stark technique, and prepared for analysis using total dis persion protocols. The pH and conductivity Were measured using a JenWay 4330 conductivity and pH meter. Anion con tent Was determined by ion chromatography using a Dionex
DX 600 series chromatograph With an Ion-Pac AS4A-SC analytical column. An inductively coupled argon plasma atomic emission spectrometer (Varian Vista RL model ICP
AES) Was used to measure 28 individual elements. Carbonate and bicarbonate content Were measured using an alkalinity titration titrator (Metrohm Titrino Model 751). i. Comparison of Maximum Experimental Centrifuge Rates
With and Without Gypsum
[0066] High throughput tests Were performed using the
Alfa Laval Lynx 1000 centrifuge (With rotating assembly #1 and rotating assembly #2) With and Without gypsum (Table lis meters (Endress & Hauser) Were used for PET and high solids slurry applications. The density of PET at the dredge and at the pilot Was measured With nuclear density meters
(Kay Ray 3680). An on-site ?eld lab Was used to conduct analyses (Table 3; ARIas required) and to collect sub samples for further lab bench analyses (Table 4).
TABLE 3
Test
Rheology bly #1 and a throughput of 54 dtph Was achieved With rotating assembly #2, When no gypsum Was added. Rotating assembly
#1 achieved 67 dtph, and rotating assembly #2 achieved 73
Flocculant Gypsum Slurry FFT Centrifuge feed Centrate Cake
Daily/AR
Daily/AR Y Y
AR
Y Y
AR
Test
OWS composition
(Dean Stark)
Methylene blue
Coulter PSD
XRD wéllt?r Ch?mlsny
MICTOSCOPY
Cold Spin
TABLE 4
Dilution
Water FFT
Centrifuge feed Centrate Cake
Y Y Y Y
AR
Y
Y
AR
AR
Y
Y
AR
AR
Y
Y
AR
AR
Y
Y
AR
AR dtph With the addition of gypsum. Gypsum addition to the
FFT feed signi?cantly improved Alfa Laval Lynx 1000 throughputs by yielding a signi?cantly stronger, more con veyable Cake~
TABLE 5
Alfa Laval Lynx 1000
Test results Without gypsum, RA #1
Throughput
(dtph)
41
54
67
73
[0063] Solids content Was measured using moisture bal ances (a Mettler-Toledo unit using an IR heating element; a polymer solutions Was determined using a Bohlin Visco 88 ii. CharacteriZation of PET
[0067] The solids content of PET dredged from a particular tailings basin at various times during a tWo and a half month

US 2014/0054231 A1 Feb. 27, 2014 period is shown in FIG. 2. As can be seen from FIG. 2, the
[0068] Data for the average mineral particle size distribu tion of four different FFT samples are shoWn in FIG. 3. It has been found that the average mineral particle size distribution of FFT is fairly consistent from basin to basin. HoWever, it is understood that variations in particle size distribution may occur from basin to basin and over time. FIGS. 4-6 shoW the changes in 44 micron, 5.5 micron, and 1.9 micron particles over about a tWo month period of time. The 44 micron portion of the solids content is very consistent, While the 5.5 and 1.9 micron fractions shoW more variations.
[0069] Tailings behavior may be attributed to clay miner als. Clay size (de?ned to be particles less than 2 microns in size) and clay minerals are strongly correlated. Methods for folloWing trends in clay concentration include use of a
laser light scattering methods, and direct quanti?cation of clay minerals using x-ray diffraction (XRD). The sedigraph method is similar in principle to the hydrometer test, Where the density of a clay suspension is monitored over time. As the coarse particles settle out, the ?uid density decreases. This decrease can be related to the particle size distribution via stokes laW and information about the ?uid viscosity. The methylene blue test involves adsorption of the methylene blue dye on the clay surfaces and is best used to quantify differ ences in clay content. The methylene blue test can be con ducted on bitumen free solids from a Dean Stark separation, or directly on the slurry suspension. XRD is useful in char acterizing the clays as minerals. Table 6 summarizes particle size data for FFT samples using various methods for clay characterization.
TABLE 6
CPA is more subject to experimental error due to dif?culties in consistent sample dispersion, and loWer signal to noise as the particle size decreases. FIG. 3 shoWs that on average, the clay content (using the CPA 1.9 or 5 .5 micron) Was higher for one set of tests compared to a second set of tests. This higher clay content results in higher than average ?occulant con sumption. Overall, the FFT had a 5.5 micron clay content ranging from 45-60%, averaging about 52%. FIG. 5 shoWs that the 5.5 micron clay content increased from 50% to 53% after dredge relocation. iii. Flocculant Make Up and Characterization
[0072] The polymer preparation unit ?rst adds Water and slices the polymer beads to several microns to increase the surface area, thereby increasing the hydration rate for the polymer. This alloWs for e?icient mixing of the mother liquor to the useable concentration. At high centrifuge feed rates, the hydration time for the polymer solution is only about 20 or 30 minutes. Inadequate polymer hydration means increased dos age requirements. Although there Was no indication of this in the testing, hydration time needs to be maximized With other more viscous or less soluble polymers. The storage tank Was a conventional oil ?eld tank, With polymer solution level maintained at about 40 m3 With stirring. Aside from polymer concentration, polymer effectiveness is affected by the degree of hydration, or the extent to Which the polymer has uncoiled in solution. Both factors are related to viscosity Which Was bration of polymer viscosity as a function of solution concen tration is shoWn in FIG. 7 for SNF Flopam 3338. The polymer viscosity folloWs the Arrhenius equation given by: nIAeEa/RT (1) energy for polymer uncoiling, R is the gas constant, and T is
Wet sieve
Dean
Stark Slurry CPA CPA CPA Sedigraph Sedigraph XRD
% Solids MB MB
Passing % % %
% % %
Passing Passing Passing
%
Passing
%
Passing
Clay
%
45 pm Solids Clay Clay 44 pm 5.5 pm 1.9 pm 44 pm 2 pm Clay
96
97
96
98
96
98
98
98
100
94
91
90
90
92
91
96
93
91
37.2
40.9
37.8
26.4
36.8
27.5
36.7
42.1
39.5
40.3
39.6
38.3
33.6
35.0
40.7
33.4
34.1
40.9
75
78
69
61
71
69
70
74
76
67
64
62
62
63
58
60
61
60
66
65
68
57
68
61
67
76
70
69
62
64
59
63
55
62
61
56
94
94
95
91
95
93
96
93
91
94
91
90
95
92
93
94
94
85
55
57
58
47
57
50
60
53
52
53
49
50
51
51
49
52
51
44
29
30
30
24
29
26
32
29
28
29
26
28
28
27
26
28
28
24
99
99
99
98
98
98
99
98
98
97
97
98
96
97
95
96
97
96
60
60
60
53
58
54
60
57
59
52
53
54
50
52
47
52
52
46
51
67
53
55
57
53
65
N/A
55
62
55
62
48
49
48
55
58
53
[0070] The consistency in the FFT feed properties over the course of testing does not alloW for an appreciation of the relationship betWeen the various analytical options When one exception of X-ray diffraction Where the uncertainty is 10% or more.
[0071] Given the strong correlations among the methods for clay determination, the CPA 5.5 micron size is preferred.
The 1.9 micron size in a laser light scattering method such as
variations in the polymer concentration can be estimated.
Using the polymer and viscosity data, FIG. 8 shoWs the plot of ln (viscosity) versus 1/T (degrees Kelvin) for the 0.2% poly mer solution. This relationship can then be used to determine a corrected viscosity by referring to the viscosity and concen tration relationship established in FIG. 7.
[0073] Polymer concentrations of 0.2 and 0.4% Were tested. FIG. 9 shoWs the histogram of polymer concentrations

US 2014/0054231 A1 Feb. 27, 2014 developed using the Arrhenius equation. 88% of the data points are Within 10% of the target 0.2% polymer, and only
17% are more concentrated than 0.2%. The average polymer concentration is 0191003. This analysis is very sensitive to changes in slope or intercept. When the intercept is changed to bring the average polymer concentration to exactly 0.2% (a change in intercept from only 13.12 to 13.04), the histogram does not change signi?cantly (FIG. 10). FIG. 11 shoWs the histogram for the 0.4% polymer, using the same slope (acti vation energy) as determined from the 0.2% polymer data, but a slope ?tted to a 0.4% polymer concentration. The histogram shoWs 0.4%:006 polymer. The increased in variability for the 0.4% polymer might be due to dif?culties in maintaining proper mixing or hydration at this higher polymer concentra tion. HoWever, the viscosity method is useful due to variations in suspended solids in the polymer make-up Water, and the dilution Water having almost 1500 ppm dissolved salts
[0074] Polymer hydration is the degree to Which the poly mer molecules have uncoiled or effectively gone into solu tion. Viscosity changes over time may be used to evaluate polymer hydration. Prior to use, the polymer Was stored in tanks With stirring Which may have helped hydrate the poly mer or break up the polymer strands in solution, resulting in viscosity changes. To ensure proper polymer hydration, a sample Was taken from the polymer solution in the storage tank and the viscosity determined. Gentle or aggressive stir ring for several minutes shoWed no change in polymer vis cosity, con?rming that the polymer Was completely hydrated.
During testing, the polymer make up Was not keeping pace
With demand, and testing commenced using 0.4% rather than mer solutions corresponded With the maximum centrifuge throughputs. At high throughputs, about 20 m3/h of the 0.4%
?occulant solution Was required. This increase in concentra tion had the effect of increasing the hydration time in the storage tank. v. Fines Capture minimum performance requirement to limit re-handling.
Fines capture is largely determined by the loss of solids in the centrate. In the ?eld, solids content determinations (e.g., bitu men, total dissolved solids, particle siZe distribution) may help guide performance. Understanding the particle siZe dis tributions in a centrifuge operation is important because of the possibility of separating ultra ?nes from the FFT. These
Would generally be the particles less than 1 micron in siZe and if they are concentrated in the centrate, there is a potential for them to create tailings handling issues far in excess of their mass fraction. This is not an issue With ?occulated FFT. The operating criteria for the ?eld solids capture Was set at 97%.
Solids capture Was the primary metric used to determine centrifuge performance in the ?eld as determined by the folloWing equation Where X is Weight percent and p is den sity:
X Capture _ Xfeed 'Pfeed — Xcentrate 'pcentrare _ Xmke - pea/(L,
Xcake 'pcake _ Xcentrate 'Pcemme Xfeed -pf”d
(2)
[0076] FIG. 12 shoWs all of the ?eld data for solids capture compared to the ?nes capture (from the laboratory analysis of
P SD), and to a clay capture determined from the average clay content of the various samples. There is a relatively loW sand content in the FFT feed since the total solids capture and the
?nes capture are almost directly correlated. Similarly, at the target ?nes capture region >95%, the clay capture is also essentially the same as the solids capture, indicating that there is no segregation of the ultra?ne solids to the centrate stream.
FIG. 13 shoWs that ultra?nes separation does not occur With
?occulated centrifuge feed, by shoWing a comparison of the siZe particle distributions for the feed, cake and centrate.
Within experimental uncertainties, these three streams have similar particle siZe distributions.
[0077] Centrate quality (suspended solids Wt %) tends to de?ne the solids or ?nes capture. FIG. 14 compares centrate solids to solids capture for three separate pilot programs over four years. As testing progressed, feWer test runs lead to off speci?cation or less than 95% capture, and as throughput increases (i.e., successively larger capacity machines Were tested), higher solids in the centrate Will still result in accept able overall ?nes or solids capture. vi. Centrate Quality
[0078] Centrate can be recycled and used to control centri fuge feed density via a dilution circuit, and may be used for polymer make up. Since polymer make up requires slicing the polymer beads into a high surface area, any solids contami nation in the preparation Water could have a deleterious effect on equipment reliability. FFT or FFT dilution, hoWever, does not require high quality Water. FIG. 14 indicates that the majority of the centrate samples contained less than 1% solids
Which Was Within an acceptable range for recycle Water in the pond and centrifuge feed dilution. vii. Centrate Settling and High FloW Rate Testing
[0079] The nominal capacity of a centrifuge depends upon the settling or separation behavior of the feed. In FFT appli cations, the ef?ciency of the separation depends upon hoW e?iciently the polymer contacts the suspension solids. The optimum polymer injection point Was found to be as close to the centrifuge as possible, implying that the polymer mixing is sensitive to overshear conditions Which might occur When polymer is injected prior to How meters and piping bends. If polymer mixing is occurring exclusively in the centrifuge, there may be high ?oW rates that overmix the polymer and
FFT. It has been previously demonstrated that centrifuge throughput is limited by lack of scroll or back drive capacity.
There might be a How rate Where overmixing prevents e?i cient separation, even With back drive capacity.
[0080] High ?oW rate runs Were conducted to assess if overmixing might make increasedback drive capacity of little or no bene?t. FIG. 15 shoWs this increasing ?oW rate experi ment and the subsequent centrate solids at those ?oW rates. As the How rate or tonnes of solids throughput increases, the centrate quality decreases. Even at the highest ?oW rates, no unusual vibrations, bearing heating, or ?uid leakages Were noted. Table 10 shoWs the 24 hour settling behavior of the centrates collected during this high volume test. Overshear or overmixing of the polymer and FFT mixture Was observed at the very highest throughput of 270 m3/h or 98 dtph, since after
24 h of settling, a signi?cant proportion of the centrate solids remained in suspension. At the loWer rates, the centrifuge feed is Well ?occulated and settles rapidly, but simply not e?iciently removed from the centrifuge. This indicates that
With properly mixed centrifuge feed, the consequences of some off speci?cation centrifuge performance Will be mini

US 2014/0054231 A1 Feb. 27, 2014 mal. These results also con?rm that increased back drive capacity can provide signi?cant improvement in centrifuge
Throughput
(dry tonnes per hour)
55
66
72
79
85
90
98
TABLE 7
% Solids in
Total FloW Centrate Solids supernatant after 24 h of
(m3/h) (%) centrate settling
160
190
210
228
245
258
271
0.51
2.03
5 .89
8.17
13.34
13.28
15.83
On spec
0.28
0.36
0.25
0.37
0.23
2.10 viii. Cake Quality
[0081] Cake properties are a function of the solids content and Water chemistry. The importance of gypsum addition in improving conveyability of the cake from the centrifuge is generally re?ected in the strength of the cake product. There is a de?nite relationship betWeen gypsum addition and cen trifuge cake strength. The ?eld laboratory used a Haake vis cometer to measure cake yield point. Table 1 1 summarizes the effect of gypsum With an average of the gypsum and non gypsum data. For the same average solids content, the gyp sum cake is considerably stronger.
Examples of additives useful in the present invention include
Portland cement, ?y ash, overburden, gypsum and lime.
[0085] From the foregoing description, one skilled in the art can easily ascertain the essential characteristics of this inven tion, and Without departing from the spirit and scope thereof, can make various changes and modi?cations of the invention to adapt it to various usages and conditions. Thus, the present invention is not intended to be limited to the embodiments shoWn herein, but is to be accorded the full scope consistent
With the claims, Wherein reference to an element in the sin gular, such as by use of the article “a” or “an” is not intended to mean “one and only one” unless speci?cally so stated, but rather “one or more”. All structural and functional equivalents to the elements of the various embodiments described throughout the disclosure that are knoWn or later come to be knoWn to those of ordinary skill in the art are intended to be encompassed by the elements of the claims. Moreover, noth ing disclosed herein is intended to be dedicated to the public regardless of Whether such disclosure is explicitly recited in the claims. a) providing a tailings feed having a solids content in a b) adding a ?occulant to the tailings feed and suf?ciently mixing the ?occulant and tailings feed to form ?ocs; and c) centrifuging the ?occulated tailings feed to produce a
Gypsum Dose
(gtonne)
0
1791
TABLE 8
Cake Yield
(Pa)
1095
12 89
Solids
(Wt %)
51.1
5 1. 1 ix. Polymer Dose, Clay Content, and Centrifuge Performance
[0082] Testing Was conducted to assess ?occulant dosages.
FIG. 16 shoWs loW ?occulant testing, all With on speci?cation
?nes capture, and the relationship betWeen throughput and
?occulant dosage. During the initial part of the test, the aver age dosage Was 962 g/tonne at 50 dtph. In the latter stages, polymer consumption Was 848 g/tonne at 36 tph throughput.
These results indicate that mixing Was probably more opti mum, possibly because the polymer injection could be located close to the centrifuge, eliminating feed tube prob lems. Coupled With the average higher clay content, polymer dosage is likely close to an optimum. At higher throughputs, polymer dosage is higher for various reasons. High gypsum dosages increases polymer requirements. At higher than pre dicted throughputs, the polymer effectiveness may also have been reduced due to loWer residence times in the hydration tank. Higher than expected tonnage throughput might also require the higher cake strength Which is associated With higher polymer dosage. It is important to note, hoWever, that there Was no explicit effort to demonstrate loWest possible
?occulant dosage at the highest tonnages.
[0083] FIG. 21 shoWs the relationship betWeen changes in clay content (both 1.9 and 5.5 micron) and polymer dosage.
Higher clay content requires an increase in polymer dosage.
With further mixing optimiZation and loW polymer dose test ing, the increase in ?occulant dosage With increased clay content is less obvious toWards the end of the test program.
[0084] In one embodiment, the centrifuge cake is further treated With an additive to give additional strength to the cake. and a cake having a solids content of at least about 50 Wt
%.
2. The process of claim 1, Wherein a coagulant is added to the tailings feed prior to the centrifuging step.
3. The process of claim 1, Wherein a coagulant is added to the tailings feed prior to the adding of the ?occulant step.
4. The process of claim 3, Wherein the coagulant and tail ings feed are mixed in an agitated feed tank for a period of time su?icient to neutraliZe repulsive electrical charges sur rounding the solids to destabiliZe suspended solids and to cause the solids to agglomerate.
5. The process of claim 3, Wherein the coagulant is added to the tailings feed in-line for a period of time suf?cient to neutraliZe repulsive electrical charges surrounding the solids to destabiliZe suspended solids and to cause the solids to
6. The process of claim 3, further comprising diluting the coagulant With Water prior to adding it to the tailings feed.
7. The process of claim 3, Wherein the dosage of coagulant ranges from about 300 grams to about 1,500 grams per tonne of solids in the tailings feed.
8. The process of claim 3, Wherein the coagulant comprises gypsum, alum or lime.
9. The process of claim 1, Wherein the solids content of the
10. The process of claim 1, Wherein in step (b), the ?occu lant and tailings feed are mixed in a static mixer having no moving parts or a dynamic mixer having at least one moving part.
11. The process of claim 10, Wherein the ?occulant and tailings feed are mixed in a static mixer having no moving parts.
12. The process of claim 11, Wherein the static mixer having no moving parts is a T mixer.
13. The process of claim 10, Wherein the ?occulant and tailings feed are mixed in a dynamic mixer having at least one

US 2014/0054231 A1 Feb. 27, 2014
14. The process of claim 13, wherein the ?occulant is added to the tailings feed in-line prior to mixing in the dynamic mixer having at least one moving part.
15. The process of claim 1, further comprising diluting the
?occulant prior to adding it to the tailings feed.
16. The process of claim 1, Wherein the ?occulant is an anionic, nonionic, cationic or amphoteric polymer.
17. The process of claim 16, Wherein the dosage of ?oc culant ranges from about 400 grams to about 2000 grams per tonne of solids in the feed.
18. The process of claim 15, Wherein the ?occulant is the form of a 0.2 to 2% by Weight aqueous solution.
19. The process of claim 15, Wherein the ?occulant is in the form of a 0.2 to 0.4% by Weight aqueous solution.
20. The process of claim 16, Wherein the ?occulant com prises a polyacrylamide anionic ?occulant.
21. The process of claim 1, Wherein in step (b), the ?occu lant is added to and mixed With the tailings feed in a centri
22. The process of claim 1, Wherein the ?occulated tailings feed is centrifuged in a solid boWl decanter centrifuge.
23. The process of claim 1, Wherein the oil sand tailings is
?uid ?ne tailings Which is optionally diluted With dilution
Water to produce the tailings feed having a solids content in
24. The process of claim 23, Wherein after step (c), the centrate is returned to a tailings pond or recycled as dilution
Water.
25. The process of claim 1, Wherein after step (d), the cake is disposed in an area using a dry stacking mode of disposal.
26. The process of claim 1, Wherein the oil sand tailings comprises ?uid ?ne tailings.
27. The process as claimed in claim 1, Wherein the process is performed at ambient temperature.
28. The process as claimed in claim 1, further comprising:
(d) deWatering the cake by consolidation, drying and/or desiccation using 1 to 2 meter lifts.
29. The process as claimed in claim 1, further comprising:
(d) deWatering the cake by consolidation, drying and/or desiccation using greater than 2 meter lifts.
30. The process as claimed in claim 1, further comprising:
(d) adding a cake strengthing additive.
31. The process as claimed in claim 30, Wherein the addi tive is selected from the group consisting of Portland cement,
?y ash, overburden, gypsum and lime.
32. The process as claimed in claim 1, Wherein the ?occu lant is a Water soluble polymer having a molecular Weight ranging betWeen about 1,000 kD to about 50,000 kD and an intrinsic viscosity of at least 3 dl/ g.
* * * * *